Hello Everyone! My name is Oscar L. Olvera Astivia.
I tend to conceptualize my research into three broad “streams” or “lines of inquiry”:
Data-generating algorithms for Monte Carlo simulations.
Computer simulations are the primary tool or methodology used by quantitative researchers when conducting psychometric or statistical research. However, to conduct well-designed, informed simulations, it is crucial to understand how to create data with the properties we wish to test. I am interested in exploring the often unacknowledged mathematical properties of these algorithms, as well as how these properties interact with published simulation results.
Non-normality in educational measurement and psychometrics.
I tend to categorize the assumptions we make when performing data analysis as either “explicit” or “implicit.” Explicit assumptions are those we find in our textbooks or articles, and we are usually very adept at determining how they can influence our inferences. Implicit assumptions are much more subtle because they become part of the data analyst’s repertoire without necessarily being mathematically or theoretically justified. I’m interested in understanding what happens when we make these distributional assumptions when they are, in fact, unwarranted.
Their impact in applied psychometrics.
Ultimately, all of these assumptions can find their way into our research, articles, reports, and recommendations. They can either alter the properties of the tests we use (like changing the sampling distribution of the parameter estimates) or they can transform the ways in which we conceptualize and understand the process of educational measurement. I’m interested in understanding what needs to be true (both in terms of epistemology and statistical theory) for the conclusions and inferences from our psychometric analyses to be warranted (i.e., valid).
For those of you who may be curious, I completed my PhD in Measurement, Evaluation and Research Methodology (MERM) in the University of British Columbia (UBC) in Vancouver, Canada. I am also an affiliated member of the Structural Equation Modeling lab in the Quantitative Methods program in the Psychology Department.
Back in the day I was also a post-doctoral research fellow in the School of Population and Public Health (in UBC) and was the lead data analyst of the Parent & Child Early Coaching Project (PACE)
psychometroscar 